Domain-Dependent Detection of Light Verb Constructions

What it is about

The paper detects light verb constructions (LVCs), such as "take a decision", and checks whether methods built for one domain carry over to another. The source domain is a Wikipedia set with 368 LVC occurrences; the target is a parallel corpus with everyday topics (magazines and novels), where the authors drop "have" and "go" from the light-verb list and add "pay" and "catch". Rule-based features are the most frequent verb, a noun suffix, and whether the noun is deverbal. A machine-learning approach is tried too. On the source domain the best single feature is most-frequent-verb at 41.94% F, but on the target it falls to 24.67%. Adapting the rules adds 1.5 F points on average, and 6.17 for the most-frequent-verb feature. Syntactic information adds 2.58 and 2.45 points on the two domains.

Ideas

• I1. A light-verb rule in en/src/expert.rs: a light verb from a short list as the head of a verb-object pair with a deverbal noun, using the intersection the paper finds most precise. Measure: F on EWT test light-verb cases, against a POS-pattern baseline, two identical runs.
• I2. Domain transfer probe: train the Mova rule on one EWT genre and test on another; the paper's drop from 41.94 to 24.67 is the size of effect to look for. Measure: the drop per genre. GUM-based sets are NC, so they are test only.
• I3. Keep the light-verb list in data/lemmas.tsv as a versioned file, and check its change with a two-run comparison, as the paper's list change shows how much the list matters.

For Mova

The chapter is 19. The lexicon: vocabulary, multiword expressions and register, with 37. Register and variation for the domain effect. The code is en/src/expert.rs (the rule) and en/src/ctx.rs (genre as context). The paper's corpora are not Mova data; the Wikipedia set is CC-licensed but NC-type here, so use it for tests only.

Cite

The paper itself, not this card. BibTeX from the library's record of the work.

@inproceedings{nagy2011domain,
  title = {{Domain-Dependent Detection of Light Verb Constructions}},
  author = {Nagy, István T. and Berend, Gábor and Móra, György and Vincze, Veronika},
  booktitle = {Proceedings of the Second Student Research Workshop associated with RANLP 2011},
  year = {2011},
  url = {https://aclanthology.org/R11-2001/}
}

References

20 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Hassan Al-Haj and Shuly Wintner. 2010. Identifying multi-word expressions by leveraging morphologi-ADAPT 31.04 43.33 36.18 42.72 37.91 40.18 43.60 38.33 40.79 47.37 37.5 41.86 46.44 40.83 43.46 49.24 40.83 44.65 cal and syntactic idiosyncrasy. In Proceedings of Coling 2010, Beijing, China. find
2. Colin Bannard. 2007. A measure of syntactic flexibility for automatically identifying multiword expressions in corpora. In Proceedings of the Workshop on a Broader Perspective on Multiword Expressions, MWE ’07, pages 1–8, Morristown, NJ, USA. ACL. find
3. Francesca Bonin, Felice Dell’Orletta, Giulia Venturi, and Simonetta Montemagni. 2010. Contrastive filtering of domain-specific multi-word terms from different types of corpora. In Proceedings of the 2010 Workshop on Multiword Expressions: from Theory to Applications, pages 77–80, Beijing, China, Aug… find
4. Coling 2010 Organizing Committee. find
5. Nicoletta Calzolari, Charles Fillmore, Ralph Grishman, Nancy Ide, Alessandro Lenci, Catherine MacLeod, and Antonio Zampolli. 2002. Towards best practice for multiword expressions in computational lexicons. In Proceedings of LREC-2002, pages 1934– 1940, Las Palmas. find
6. Laura Chiticariu, Rajasekar Krishnamurthy, Yunyao Li, Frederick Reiss, and Shivakumar Vaithyanathan. 2010. Domain adaptation of rule-based annotators for named-entity recognition tasks. In Proceedings of EMNLP 2010, pages 1002–1012, Stroudsburg, PA, USA. ACL. find
7. Paul Cook, Afsaneh Fazly, and Suzanne Stevenson. 2007. Pulling their weight: exploiting syntactic forms for the automatic identification of idiomatic expressions in context. In Proceedings of the Workshop on a Broader Perspective on Multiword Expressions, pages 41–48, Morristown, NJ, USA. ACL. find
8. Hal Daumé III. 2007. Frustratingly easy domain adaptation. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, pages 256–263, Prague, Czech Republic, June. Association for Computational Linguistics. find
9. Afsaneh Fazly and Suzanne Stevenson. 2007. Distinguishing Subtypes of Multiword Expressions Using Linguistically-Motivated Statistical Measures. In Proceedings of the Workshop on A Broader Perspective on Multiword Expressions, pages 9–16, Prague, Czech Republic. Association for Computational Linguis… find
10. Su Nam Kim. 2008. Statistical Modeling of Multiword Expressions. Ph.D. thesis, University of Melbourne, Melbourne. find
11. Dan Klein and Christopher D. Manning. 2003. Accurate unlexicalized parsing. In Proceedings of the 41st Annual Meeting on Association for Computational Linguistics, ACL ’03, pages 423–430, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
12. Andrew Kachites McCallum. 2002. Mallet: A machine learning for language toolkit. http://mallet.cs.umass.edu. ↗
13. Carlos Ramisch, Aline Villavicencio, and Christian Boitet. 2010a. Multiword Expressions in the wild? The mwetoolkit comes in handy. In Coling 2010: Demonstrations, Beijing, China. find our card
14. Carlos Ramisch, Aline Villavicencio, and Christian Boitet. 2010b. mwetoolkit: a framework for multiword expression identification. In Nicoletta Calzolari, Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, and Daniel Tapias, editors, Proceedings of LREC’10, Valletta, Malta… find
15. Carlos Ramisch, Aline Villavicencio, and Christian Boitet. 2010c. Web-based and combined language models: a case study on noun compound identification. In Coling 2010: Posters, Beijing, China. find
16. Ivan A. Sag, Timothy Baldwin, Francis Bond, Ann Copestake, and Dan Flickinger. 2002. Multiword Expressions: A Pain in the Neck for NLP. In Proceedings of CICLing-2002, pages 1–15, Mexico City, Mexico. find our card
17. György Szarvas, Richárd Farkas, and András Kocsor. 2006. A Multilingual Named Entity Recognition System Using Boosting and C4.5 Decision Tree Learning Algorithms. In Discovery Science, pages 267–278. find
18. Kristina Toutanova and Christopher D. Manning. 2000. Enriching the knowledge sources used in a maximum entropy part-of-speech tagger. In Proceedings of EMNLP 2000, pages 63–70, Stroudsburg, PA, USA. Association for Computational Lingustics. find
19. Veronika Vincze and János Csirik. 2010. Hungarian corpus of light verb constructions. In Proceedings of Coling 2010, Beijing, China. find
20. Veronika Vincze, István Nagy T., and Gábor Berend. 2011. Detecting noun compounds and light verb constructions: a contrastive study. In Proceedings of the Workshop on Multiword Expressions: from Parsing and Generation to the Real World, pages 116– 121, Portland, Oregon, USA, June. Association for… find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2011-nagy-domain-dependent-detection-light-verb/